-
Notifications
You must be signed in to change notification settings - Fork 80
/
train_PL.py
202 lines (179 loc) · 10.1 KB
/
train_PL.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
import os
import time
import argparse
import sys
import numpy as np
import torch
import torch.optim as optim
from tqdm import tqdm
from network.BEV_Unet import BEV_Unet
from network.ptBEV import ptBEVnet
from dataloader.dataset import collate_fn_BEV,spherical_dataset,voxel_dataset
from dataloader.dataset_PL import coarseID_name,PLfine2coarse,PLY_dataset
from network.lovasz_losses import lovasz_softmax
#ignore weird np warning
import warnings
warnings.filterwarnings("ignore")
def fast_hist(pred, label, n):
k = (label >= 0) & (label < n)
bin_count=np.bincount(
n * label[k].astype(int) + pred[k], minlength=n ** 2)
return bin_count[:n ** 2].reshape(n, n)
def per_class_iu(hist):
return np.diag(hist) / (hist.sum(1) + hist.sum(0) - np.diag(hist))
def fast_hist_crop(output, target, unique_label):
hist = fast_hist(output.flatten(), target.flatten(), np.max(unique_label)+1)
hist=hist[unique_label,:]
hist=hist[:,unique_label]
return hist
def SemKITTI2train(label):
if isinstance(label, list):
return [SemKITTI2train_single(a) for a in label]
else:
return SemKITTI2train_single(label)
def SemKITTI2train_single(label):
return label - 1
def main(args):
data_path = args.data_dir
train_batch_size = args.train_batch_size
val_batch_size = args.val_batch_size
check_iter = args.check_iter
model_save_path = args.model_save_path
compression_model = args.grid_size[2]
grid_size = args.grid_size
pytorch_device = torch.device('cuda:0')
model = args.model
if model == 'polar':
fea_dim = 9
circular_padding = True
elif model == 'traditional':
fea_dim = 7
circular_padding = False
#prepare miou fun
unique_label=np.asarray(list(coarseID_name.keys()))
unique_label_str=np.asarray(list(coarseID_name.values()))
#prepare model
my_BEV_model=BEV_Unet(n_class=len(unique_label), n_height = compression_model, input_batch_norm = True, dropout = 0.2, circular_padding = circular_padding)
my_model = ptBEVnet(my_BEV_model, pt_model = 'pointnet', grid_size = grid_size, fea_dim = fea_dim, max_pt_per_encode = 256,
out_pt_fea_dim = 512, kernal_size = 1, pt_selection = 'random', fea_compre = compression_model)
if os.path.exists(model_save_path):
my_model.load_state_dict(torch.load(model_save_path))
my_model.to(pytorch_device)
optimizer = optim.Adam(my_model.parameters())
loss_fun = torch.nn.CrossEntropyLoss(ignore_index=255)
#prepare dataset
train_PLY_dataset1 = PLY_dataset('../data/paris_lille/Lille2.ply',0.1,0.05,label_convert_fun = PLfine2coarse,return_ref=True)
train_PLY_dataset2 = PLY_dataset('../data/paris_lille/Lille1.ply',0.1,0.05,label_convert_fun = PLfine2coarse,return_ref=True)
train_PLY_dataset = torch.utils.data.ConcatDataset([train_PLY_dataset1,train_PLY_dataset2])
val_PLY_dataset = PLY_dataset('../data/paris_lille/Paris.ply',0.1,0.05,label_convert_fun = PLfine2coarse,return_ref=True)
if model == 'polar':
train_dataset=spherical_dataset(train_PLY_dataset, grid_size = grid_size, flip_aug = True, ignore_label = 255, rotate_aug = True, fixed_volume_space = True,\
max_volume_space = [15,np.pi,12],min_volume_space = [0,-np.pi,-3])
val_dataset=spherical_dataset(val_PLY_dataset, grid_size = grid_size, ignore_label = 255, fixed_volume_space = True, max_volume_space = [15,np.pi,12],min_volume_space = [0,-np.pi,-3])
elif model == 'traditional':
train_dataset=voxel_dataset(train_PLY_dataset, grid_size = grid_size, flip_aug = True, ignore_label = 255, rotate_aug = True, fixed_volume_space = True,\
max_volume_space = [15,15,12],min_volume_space = [-15,-15,-3])
val_dataset=voxel_dataset(val_PLY_dataset, grid_size = grid_size, ignore_label = 255, fixed_volume_space = True,max_volume_space = [15,15,12],min_volume_space = [-15,-15,-3])
train_dataset_loader = torch.utils.data.DataLoader(dataset = train_dataset,
batch_size = train_batch_size,
collate_fn = collate_fn_BEV,
shuffle = True,
num_workers = 4)
val_dataset_loader = torch.utils.data.DataLoader(dataset = val_dataset,
batch_size = val_batch_size,
collate_fn = collate_fn_BEV,
shuffle = False,
num_workers = 4)
# training
epoch=0
best_val_miou=0
start_training=False
my_model.train()
global_iter = 0
exce_counter = 0
while True:
loss_list=[]
pbar = tqdm(total=len(train_dataset_loader))
for i_iter,(_,train_vox_label,train_grid,_,train_pt_fea) in enumerate(train_dataset_loader):
# validation
if global_iter % check_iter == 0:
my_model.eval()
hist_list = []
val_loss_list = []
with torch.no_grad():
for i_iter_val,(_,val_vox_label,val_grid,val_pt_labs,val_pt_fea) in enumerate(val_dataset_loader):
# val_vox_label = SemKITTI2train(val_vox_label)
# val_pt_labs = SemKITTI2train(val_pt_labs)
val_pt_fea_ten = [torch.from_numpy(i).type(torch.FloatTensor).to(pytorch_device) for i in val_pt_fea]
val_grid_ten = [torch.from_numpy(i[:,:2]).to(pytorch_device) for i in val_grid]
val_label_tensor=val_vox_label.type(torch.LongTensor).to(pytorch_device)
predict_labels = my_model(val_pt_fea_ten, val_grid_ten)
loss = lovasz_softmax(torch.nn.functional.softmax(predict_labels).detach(), val_label_tensor,ignore=255) + loss_fun(predict_labels.detach(),val_label_tensor)
predict_labels = torch.argmax(predict_labels,dim=1)
predict_labels = predict_labels.cpu().detach().numpy()
for count,i_val_grid in enumerate(val_grid):
hist_list.append(fast_hist_crop(predict_labels[count,val_grid[count][:,0],val_grid[count][:,1],val_grid[count][:,2]],val_pt_labs[count],unique_label))
val_loss_list.append(loss.detach().cpu().numpy())
my_model.train()
iou = per_class_iu(sum(hist_list))
print('Validation per class iou: ')
for class_name, class_iou in zip(unique_label_str,iou):
print('%s : %.2f%%' % (class_name, class_iou*100))
val_miou = np.nanmean(iou) * 100
del val_vox_label, val_grid, val_pt_fea, val_grid_ten, val_pt_fea_ten, val_label_tensor, predict_labels
# save model if performance is improved
if best_val_miou<val_miou:
best_val_miou=val_miou
torch.save(my_model.state_dict(), model_save_path)
print('Current val miou is %.3f while the best val miou is %.3f' %
(val_miou,best_val_miou))
print('Current val loss is %.3f' %
(np.mean(val_loss_list)))
if start_training:
print('epoch %d iter %5d, loss: %.3f\n' %
(epoch, i_iter, np.mean(loss_list)))
print('%d exceptions encountered during last training\n' %
exce_counter)
exce_counter = 0
loss_list = []
# training
try:
# train_vox_label = SemKITTI2train(train_vox_label)
train_pt_fea_ten = [torch.from_numpy(i).type(torch.FloatTensor).to(pytorch_device) for i in train_pt_fea]
train_grid_ten = [torch.from_numpy(i[:,:2]).to(pytorch_device) for i in train_grid]
train_vox_ten = [torch.from_numpy(i).to(pytorch_device) for i in train_grid]
point_label_tensor=train_vox_label.type(torch.LongTensor).to(pytorch_device)
# forward + backward + optimize
outputs = my_model(train_pt_fea_ten,train_grid_ten)
loss = lovasz_softmax(torch.nn.functional.softmax(outputs), point_label_tensor,ignore=255) + loss_fun(outputs,point_label_tensor)
loss.backward()
optimizer.step()
loss_list.append(loss.item())
# zero the parameter gradients
optimizer.zero_grad()
del train_pt_fea_ten, train_grid_ten, train_vox_ten, point_label_tensor, outputs, loss
except Exception:
exce_counter += 1
pbar.update(1)
start_training=True
global_iter += 1
pbar.close()
epoch += 1
if __name__ == '__main__':
# Training settings
parser = argparse.ArgumentParser(description='')
parser.add_argument('-d', '--data_dir', default='data')
parser.add_argument('-p', '--model_save_path', default='./PL_PolarSeg.pt')
parser.add_argument('-m', '--model', choices=['polar','traditional'], default='polar', help='training model: polar or traditional (default: polar)')
parser.add_argument('-s', '--grid_size', nargs='+', type=int, default = [320,320,32], help='grid size of BEV representation (default: [320,320,32])')
parser.add_argument('--train_batch_size', type=int, default=2, help='batch size for training (default: 2)')
parser.add_argument('--val_batch_size', type=int, default=2, help='batch size for validation (default: 2)')
parser.add_argument('--check_iter', type=int, default=2000, help='validation interval (default: 2000)')
args = parser.parse_args()
if not len(args.grid_size) == 3:
raise Exception('Invalid grid size! Grid size should have 3 dimensions.')
print(' '.join(sys.argv))
print(args)
main(args)